The effectiveness of compression ultrasonography in the detection of femoral and popliteal venous thrombosis was determined in a prospective trial over a period of seven months. Forty-two patients who had an isolated intertrochanteric or femoral neck fracture participated, and forty of these patients completed the study. The average age of the patients was 81.6 years. Venography was the standard with which all ultrasonic studies were compared. A total of 112 ultrasonic studies were performed. The incidence of major venous thrombosis, as detected by venography, was 12.5 per cent. All patients were clinically asymptomatic. The compression ultrasonic technique had an accuracy of 97 per cent, a sensitivity of 100 per cent, and a specificity of 97 per cent. Compression ultrasonography appears to be a very effective technique for diagnosing venous thrombosis in patients who have a fracture of the hip. It is safe, well accepted by both patients and staff, and simply and quickly performed, and it carries no inherent risks. It also can be readily repeated, making it ideal for monitoring high-risk patients.